open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi !

I'm trying to upgrade models that run well with the mascaret version 5.1.4 with the old executable on the mascaret v5.1.10 with the new executable. And I have some troubles to run some models with geometric weirs. If they are not activated, the model run well in steady (he bugs in unsteady because of the absence of the weirs what it's normal), and when i active them, i have this message :

At line 531 of file C:\opentelemac-mascaret\RD_issue\dev_schapi\plug_mascaret\sources\mascaret\Pretrait\lec_sing.f90
Fortran runtime error: End of file

Error termination. Backtrace:

Could not print backtrace: libbacktrace could not find executable to open
#0 0xffffffff
#1 0xffffffff
#2 0xffffffff
#3 0xffffffff
#4 0xffffffff
#5 0xffffffff
#6 0xffffffff
#7 0xffffffff
#8 0xffffffff
#9 0xffffffff
#10 0xffffffff
#11 0xffffffff
#12 0xffffffff
#13 0xffffffff
#14 0xffffffff
#15 0xffffffff
#16 0xffffffff

Notice these models were created initialy with the mascaret version 2 and they were upgraded until the mascaret version 5.1.4 all along these years.
Models with geometric weirs created with mascaret v5.0.7 or v5.1.4 run well, even if they are upgraded on the mascaret version 5.1.10 with the new executable. I clearly don't understand where the problem comes from. I tryed to delete these weirs and created news but i still have the problem.

I downloaded here one my problematic models.

Regards

Hi Mathieu,

It looks like a problem with allocation.
There is probably a bug as it should be able to read your geometric weir.
v5.1.10 of the plugin use an official and recent version of mascaret (~v8.4)
Old version of the plugin use unofficial and older version of mascaret.

What is surprising is that your model was OK with v5.0.8 or v5.1.4 as I expected these versions of the plugin to work with v8.2 of Mascaret (modification already done).

Anyway, for the bug : need to see that with the developers.

To fix your problem (short term):
- Simplify the geometry of your weir! I know it's easier to build your profile from raw data (DTM or topo) but it is counterproductive.
- On hydraulic aspects, it doesn't make sense. I know geometric weir is poorly documented.

For example with your geometric weir "OH50", made of 2254 points, Mascaret will see it as 2253 rectangular weirs in //.
-> Even if you define it on your desactivated profile, the concept of minor bed doesn't exist for geometric weir.
=> The weir allows water to pass through at 104.6m.
Modeling multiple weirs in case of difluence without modeling explicitely the diffluence requires more abstraction (we can discuss it by mp).

Alexis

Hi Alexis!
I simplified my geometric weirs and it works fine now !

Hi mathieu,

The number of characters per line of the xcas file is limited to 8192, unlike the previous version of the executable which was slightly higher. Therefore, simplifying the weirs geometries allows to solve the problem.

Regards,